Systematic Review of Cost-Effectiveness Analyses of Treatments for Psoriasis Wei Zhang , Nazrul Islam , Canice Ma , Aslam H

Home , Tacalcitol

Authors' [Accepted] Version

Systematic Review of Cost-Effectiveness Analyses of Treatments for Psoriasis

Wei Zhang1,2, Nazrul Islam1,2, Canice Ma1, Aslam H. Anis1,2

PharmacoEconomics. 2015; 33(4):327-40

The final publication is available at Springer via http://dx.doi.org/10.1007/s40273-014-0244-9

1Centre for Health Evaluation and Outcome Sciences, St. Paul’s Hospital, 588-1081 Burrard

Street, Vancouver, BC, V6Z 1Y6, Canada

2School of Population and Public Health, University of British Columbia, 2206 East Mall

Vancouver, BC, V6T 1Z3, Canada

Corresponding Author

Aslam H. Anis PhD, FCAHS

Centre for Health Evaluation and Outcome Sciences

St. Paul’s Hospital

588-1081 Burrard Street, Vancouver, BC V6Z 1Y6

Tel: +1-604-806-8712

Fax: +1-604-806-8778

Email: [email protected] Abstract

Objectives: To systematically review the current literature on cost-effectiveness analysis (CEA) of existing treatment options for psoriasis, assess the quality of these studies, and summarize the evidence on the drivers of cost-effectiveness.

Methods: A literature search using Medical Subject Headings and keywords was performed in the databases, MEDLINE, EMBASE, NHS EED and HTA; the CEA Registry was searched using keywords only. All references within the relevant review articles were examined manually.

Two researchers independently determined the final articles and a third researcher resolved any discrepancies. We evaluated study quality in terms of the study perspective, effectiveness measures, cost measures, economic model, and time horizon. Any sensitivity analyses conducted in the studies were examined to identify the drivers of cost-effectiveness, which are any variables leading to changes in the study conclusions.

Results: Fifty-three articles were included in our final review: about 70% did not consider costs related to adverse events; approximately one-quarter used Quality-Adjusted Life Years; 34% were short-term analyses with a time horizon under one year. Thirty-eight studies conducted a sensitivity analysis, in 18 of which could have a potential impact on the final cost-effectiveness results due to uncertain variables. The key drivers of cost-effectiveness included the costs related to the treatment, efficacy, utility values, and hospitalization for non-responders.

Conclusions: High quality cost-effectiveness studies are required to facilitate resource allocation decision-making. To improve study quality, future research should provide evidence on the long-

2 term experience with psoriasis treatments and resolve the uncertainty associated with key s of cost-effectiveness.

Key Points for Decision Makers

 Most cost-effectiveness analyses of psoriasis treatments are of low quality, with short

time horizons, non-comparable effectiveness measures, incomplete cost measures, or no

sensitivity analysis.

 Cost related to treatment, hospitalization for non-responders, efficacy, and utility values

are the key drivers of cost-effectiveness of psoriasis treatments.

 Future studies are required to provide evidence on the long-term experience with

psoriasis treatments and resolve the uncertainty associated with key drivers of cost-

effectiveness.

3 1. Introduction

Psoriasis is a common, chronic inflammatory disease of the skin. The prevalence of psoriasis in adults varies from 0.91% (United States) to 8.5% (Norway) and its incidence varies from

78.9/100,000 person-years (United States) to 230/100,000 person-years (Italy) [1]. Psoriasis has a major impact on patients’ physical and mental function, as well as their quality of life [2].

Furthermore, psoriasis is associated with increased health care resource use and costs [3].

Psoriasis mostly affects Caucasians [4], typically with onset between the ages of 15 and 30 years and with comparable frequency among males and females [5]. Despite a wide range of treatment options, including topical therapy, phototherapy, traditional systemic therapy, and biologic agents, psoriasis remains incurable, with relapses and remissions [6]. First-line therapies, conventionally prescribed for mild psoriasis, are topical agents such as emollients, tar, steroids, and vitamin D analogues. Second-line therapies, conventionally prescribed for moderate-to- severe psoriasis, include phototherapy and traditional systemic agents such as Methotrexate,

Cyclosporine, and Acitretin. Patients who fail to achieve response with, or are ineligible for, both first- and second-line therapies are prescribed biologic agents (such as Etanercept, Infliximab,

Ustekinumab, Adalimumab), which were introduced in 2003 [6–8].

Cost-effectiveness analysis (CEA) is widely used to assist decision makers to select the most appropriate treatments within constrained health care system budgets. A great number of CEAs on the various treatments of psoriasis have been published in the medical literature in recent years. To ensure resource allocation decisions are informed by high quality evidence, it is critical to assess these economic evaluations. While several systematic reviews have been conducted on

previously published CEAs [9,10], their focus has been on only biologic therapies [9] or subsequent treatments after failure of the first biologic [10]. There has been no comprehensive assessment of the quality of these CEAs and key drivers of cost-effectiveness (i.e., variables leading to changes in study conclusions) have not been identified. This study systematically reviewed all CEAs of psoriasis treatments, examined the quality of these studies, and summarized the evidence on drivers of cost-effectiveness.

2. Methods

2.1. Literature Search

A systematic literature search using the databases MEDLINE, EMBASE, NHS EED and HTA

(produced by the University of York Centre for Reviews and Dissemination), and the CEA

Registry was performed to identify all CEAs of psoriasis treatments until November 13, 2013.

The NHS EED search filter [11] was applied to identify all economic evaluations in MEDLINE and EMBASE (Appendix-I). The keywords and Medical Subject Headings (MeSH) for psoriasis were used to capture relevant economic evaluations in all databases except the CEA Registry, in which only keywords were used. In addition, we manually examined all references within the relevant review articles to identify potentially overlooked studies. Although no language exclusions were applied, only those published in English were included in the final review.

2.2 Eligibility Criteria

Study inclusion was determined by two researchers independently (NI and CM). During the first stage, titles and abstracts of all articles identified by the literature search were screened. Articles were excluded if the study was not about psoriasis (psoriatic arthritis was excluded), psoriasis

treatment, or economic evaluation, or if they were conference abstracts, letters/editorials, conference reports, errata, or book chapters. Next, the full texts of all articles deemed eligible from the previous stage were reviewed to further confirm their eligibility. At this stage, the exclusion criteria were burden of illness studies, cost analyses, study types other than economic evaluations, review articles, editorials or reports, or articles not written in English. Any discrepancies between the two reviewers were resolved by consensus with input from a third reviewer (WZ).

2.3. Data Extraction and Qualitative Synthesis

For each study included in the final review, data were extracted into a table. The information extracted included authors, country, study year, comparators, population characteristics, study design, time horizon, perspective, data sources, effectiveness, cost, sensitivity analysis, and conclusions. A qualitative synthesis was performed and the studies were summarized according to different treatment types.

2.3.1. Quality Assessment

The quality of the cost-effectiveness studies was first independently assessed by two reviewers

(NI and CM) using the Quality of Health Economic Studies (QHES) instrument developed by

Chiou et al. [12]. This scheme took into consideration Drummond’s Checklist and the US Public

Health Service Panel on Cost-Effectiveness in Health and Medicine. Although there are other instruments that can evaluate the quality of health economic studies, such as the British Medical

Journal checklist [13], the Canadian Guidelines [14], and the Journal of the American Medical

Association (JAMA) user’s guide [15,16], the QHES instrument was used because it is the only

validated instrument [17]. Consisting of 16 questions, this instrument was designed to quickly assess the common types of health economic evaluations including cost-effectiveness and cost- utility analyses [18]. For each question, a rating of “yes” or “no” was assigned by the two reviewers independently. Once again, the two reviewers discussed the discrepancies until consensus was reached with input from WZ. Data were presented as the proportion of articles reporting each of the 16 items. Furthermore, we evaluated the quality in terms of the study perspective, effectiveness measures and their data source, cost measures, economic model, and time horizon.

2.3.2. Sensitivity Analysis and Drivers of Cost-effectiveness

For each article included in the final review, we examined whether a sensitivity analysis was executed. Sensitivity analysis in CEAs is key to identifying the drivers of cost-effectiveness, as it identifies which variables may potentially change study conclusions. If a sensitivity analysis was conducted, we captured what uncertain parameter estimates were considered, their impact on the study conclusions, and what drivers of cost-effectiveness were identified.

3. Results

3.1. Literature Search

The literature search identified 2,617 abstracts: 695 from MEDLINE, 1,787 from EMBASE, 120 from NHS EED and HTA, 15 from CEA Registry, and 10 from the hand search of the references from relevant review articles (Figure 1). After the initial screening of all titles and abstracts, 500 articles were selected for full-text review. A total of 53 articles were included in the final review

after excluding burden of illness and cost analysis studies, non-economic evaluation studies, reviews, editorials, reports, non-English studies, and non-downloadable articles.

Most of the studies included in the review were from North America [19–40] and Europe [41–

67], with one multisite study across North America and Europe [68] and one each from India

[69], Israel [70], and Japan [71]. More than three-quarters (77.8%) of the studies were published after 2003, in the post-biologic era.

3.2. Quality Assessment

The majority of the studies reported objectives, sources of variable estimates, methodology of data abstraction, measurement of costs, and primary outcome measures (Table 1). However, approximately 40% of the studies only estimated the cost-effectiveness ratios (e.g., cost per responder) for each treatment option but did not undertake an incremental analysis of the treatment alternatives. About 60% of the CEA studies applied a time horizon that was truncated, meaning they did not allow time for all relevant and important outcomes to be observed. For example, discontinuation or failure of treatment, potential adverse events, or hospitalization over time were often excluded as they were not likely to occur within the short time horizon adapted for the CEA. The measurement of costs was not appropriate in approximately 40% of the studies, either because they considered only the costs of the specific treatment options or because they disregarded the costs related to adverse events (please see Section 3.3.3). More than 30% did not explicitly mention the source of funding for their study and approximately one in four studies discussed potential biases.

3.3. Synthesis of CEA Studies

Data on CEAs extracted from the selected studies are summarized in Table 2. About one-third (n

= 17) of the studies compared the cost-effectiveness of biologic therapies while only about a quarter (n = 13) were on topical therapies only. A little over one-third (n = 18) compared therapies from more than one group of psoriasis treatment modalities described above.

3.3.1. Perspective

More than half of studies (n = 29) employed a health system perspective (including the government, national health system, third-party payer, insurer, or managed care perspective).

Nine (9) studies applied a broader societal perspective and 14 studies did not explicitly state the considered perspective.

3.3.2. Effectiveness

The effectiveness measures used in the various analyses included the Psoriasis Area and Severity

Index (PASI) (n = 26), disease/therapy-free days/years (n = 7), and others such as clearing/success rates (n = 4). The Quality-Adjusted Life Years (QALY) gained was estimated and used in about a quarter (n = 14) of the studies reviewed. Another study conducted a CEA using utilities assessed by a visual analog scale and a cost-benefit analysis using the willingness- to-pay method. Efficacy evidence was estimated by a systematic review of the literature, which includes meta-analyses, in 32 of the studies.

3.3.3. Cost

The costs considered in the analyses were medication (including phototherapy, as applicable) costs only (n = 9), direct health care costs without considering adverse events (n = 19), direct costs including the costs of adverse events (n = 13), direct and indirect (for example, absenteeism/unemployment) costs (n = 6), and direct and indirect costs plus the costs of adverse events (n = 6). In terms of the indirect costs, only one article took account of presenteeism, which is reduced productivity while working due to health problems.

3.3.4. Economic Model and Time Horizon

Decision analytic and Markov models were used in 13 and 12 of the studies respectively. Four studies used the “York” model developed by Woolacott et al. [67]; 9 studies used trial-based analysis, and 7 studies did not specify using any models. The CEA studies also varied in time horizon: less than 1 year in 18 studies, between 1 and 3 years in 25 studies, and more than 3 years in 7 studies.

3.3.5. Final Decisions

The study design, analytic methods, and comparators used were diverse across the studies; therefore determining a single conclusion based on these findings is difficult. Despite these variations, some of the agents that were found to be cost-effective were Adalimumab (n = 7),

Methotrexate (n = 6), Infliximab (n = 5), Etanercept (n = 5), Ustekinumab (n = 4), Calcipotriol (n

= 3), Cyclosporine (n = 2), Phototherapy (n = 2), and fixed combination of

Calcipotriol/Betamethasone (n = 2).

3.4. Drivers of Cost-effectiveness

A total of 38 studies conducted sensitivity analyses (Table 3). Among them, 25 studies conducted one-way and/or two-way sensitivity analysis to measure the impact on final outcomes when one variable or two variables (simultaneously) were changed while all other variables were held constant. One study conducted a multi-way sensitivity analysis only, by simultaneously changing multiple variables. The remaining 12 studies conducted probabilistic sensitivity analysis, either alone or with one-way, two-way, or multi-way sensitivity analysis. The most common uncertain variables considered for the analysis were cost and efficacy of the treatment.

We found that in 20 studies, the uncertainty analyses had no impact on the final conclusions of the study whereas in the other 14 studies, the cost-effectiveness conclusions were definitely affected. In the remaining 4 studies, the authors either did not clearly state the threshold for final cost-effectiveness decision or present the impact results of sensitivity analyses. In these studies, there is a potential impact of uncertain variables on the cost-effectiveness conclusions.

According to the 18 studies where we found a definite impact or a potential impact of uncertain variables on the final conclusions, the key drivers included: (1) the costs of the medication or phototherapy itself, or related factors such as dosage, treatment-free response period, average wholesale price, and weight [21,22,24,27,34,43,46,50,51,71]; (2) the overall costs including medication, physician, and laboratory costs [34,61]; (3) efficacies including PASI response rate,

Dermatology Life Quality Index response rate, and clearing rate [23,26,27,34,43,46,59,61]; (4) utility values, which comprised of those for different disease severity levels or PASI response health states, baseline utility values, and utility ratings for the side effects or being on waiting list

[22,26,51,57,67]; (5) hospitalization assumptions for non-responders [24,62,67]; (6) others including lost productivity while hospitalized [62] and different time horizons [40].

4. Discussion

This is the first article that searched for and reviewed the available CEA studies for all treatment options in psoriasis. We focused on assessing the quality of these studies and identifying the drivers of cost-effectiveness instead of the actual cost-effectiveness outcomes. We conclude the

CEA studies of high quality should apply a reasonably long time horizon, adopt a valid and comparable effectiveness measure, consider all cost items relevant to the study perspective, and conduct a sensitivity analysis to assess the uncertainty around parameter estimates. We found only a small number of the studies met these rigorous standards [22,35,40,43,51,53,56–59,62,67].

Over 30% of CEAs of psoriasis treatments adopted a time horizon of less than one year, despite the fact that psoriasis is a chronic disease with negative impacts on health and quality of life.

Although a long time horizon is preferred, information related to the long-term experience with various psoriasis treatments is largely lacking [10,67]. This includes the annual drop-out rates from therapy, the ‘remission’ period assumed between spells of intermittent treatment, the efficacy of subsequent lines of treatment, the cost and incidence of adverse events, and the risk of hospitalization [10,67].

A great number of these CEAs evaluated the cost per responder and the PASI was often used to identify treatment response. Regardless of its limitations [72] and other recommendations on alternative weighting in scoring [73], the PASI has been found to be reliable and valid [74].

Some researchers proposed using an alternative scoring system called the Psoriasis Assessment

Severity Score (PASS) as it is more sensitive than the PASI [75]; however, the US Food and

Drug Administration (FDA) requires PASI scores to be reported upon evaluating efficacy of new therapeutic modalities of psoriasis [76]. While PASI-75 (a 75% reduction in the PASI score) is the current benchmark in reporting the primary endpoints of psoriasis trials, PASI-50 (a 50% reduction) has also been found to be a clinically significant endpoint [77]. A variety of cut-offs have been used in the reviewed studies (PASI-50, PASI-75, and PASI-90). However, having only a disease-specific outcome such as cost per PASI-75 responder in CEAs, we could not determine the cost-effectiveness of an intervention due to the lack of a commonly accepted threshold or compare the cost-effectiveness results to cost-effectiveness found in other diseases.

QALY has been widely used and identified as the most important effectiveness measure to employ in CEAs over the last decade or so [78]. The National Health Service in England and

Wales uses QALY as the principal measure of health outcome as recommended by the National

Institute of Health and Clinical Excellence [78], though we found that about three-quarters of the studies reviewed did not use this measurement.

When estimating direct health care costs, potential adverse events were often ignored. Only one study considered adverse events as well as the impact of treatment on work productivity including absenteeism and presenteeism. Previous studies [79,80] have found that having psoriasis was associated with substantial work productivity loss in terms of both absenteeism and presenteeism. Furthermore, indirect costs due to productivity loss could exceed the direct health care costs of psoriasis treatments. It was suggested that future effectiveness and cost- effectiveness studies on new psoriasis treatments should take into account the new interventions’ impact on patient productivity and the corresponding economic burden [79,80]. Both absenteeism and presenteeism were ignored in most of the CEA studies. The omission is

primarily due to the lack of empirical data on the work productivity impact of treatment. Future studies should incorporate the work productivity measure into the cost-effectiveness study model.

Sensitivity analysis was conducted in 38 studies. In over half of these studies, we did not find any impact of the uncertain parameter estimates of interest on the final conclusions. For the studies where we found an impact, the costs related to the treatment, treatment efficacy, utility values, and hospitalization for non-responders were the key drivers for the cost-effectiveness conclusions. Our findings may be subject to publication bias because studies may not be published if their results were highly sensitive to the uncertain parameter estimates.

One limitation of our review is that we may have missed a few CEA studies. Several CEAs were conducted directly by manufacturers [67,81–83] but the details were not published in literature; therefore they were not accessible as our search was restricted to literature publications only. In addition, we limited the final review to studies in English. Furthermore, there were 2 articles that could not be downloaded because they were studies from the early 1990s. Despite these drawbacks, we included the majority of CEA studies in literature, which enabled us to evaluate the overall quality and identify key drivers of cost-effectiveness.

In conclusion, high quality cost-effectiveness studies are needed to inform resource use decision- making for psoriasis treatments. To improve the quality, further studies should be conducted to provide evidence related to the long-term experience with different treatments and to address the uncertainty associated with the key drivers of cost-effectiveness.

ACKNOWLEGEMENTS

Wei Zhang is a recipient of the Michael Smith Foundation for Health Research Postdoctoral

Fellowship Award. Nazrul Islam acknowledges support from the Canadian Institutes of Health

Research through the Vanier Canada Graduate Scholarship. Wei Zhang, Nazrul Islam, and

Canice Ma declare no conflicts of interest regarding the formation of this review.

Aslam Anis reports receiving grants from AbbVie Corporation and Pfizer Inc. outside the submitted work, as well as personal fees from AbbVie Corporation, Pfizer Inc., International

Federation of Psoriasis Association, and International Value Coalition (via the European

Accreditation Council for Continuing Medical Education).

References

1. Parisi R, Symmons DP, Griffiths CE, Ashcroft DM. Global epidemiology of psoriasis: a systematic review of incidence and prevalence. J Invest Dermatol. 2013;133:377–85.

2. Rapp SR, Feldman SR, Exum ML, Fleischer AB Jr, Reboussin DM. Psoriasis causes as much disability as other major medical diseases. J. Am. Acad. Dermatol. 1999;41:401–7.

3. Yu AP, Tang J, Xie J, Wu EQ, Gupta SR, Bao Y, et al. Economic burden of psoriasis compared to the general population and stratified by disease severity. Curr. Med. Res. Opin. 2009;25:2429–38.

4. Lowes MA, Bowcock AM, Krueger JG. Pathogenesis and therapy of psoriasis. Nature. 2007;445:866–73.

5. Gudjonsson JE, Elder JT. Psoriasis: epidemiology. Clin. Dermatol. 2007;25:535–46.

6. Dvorakova V, Markham T. Psoriasis: current treatment options and recent advances. Prescriber. 2013;24:13–20.

7. Rich SJ, Bello-Quintero CE. Advancements in the treatment of psoriasis: role of biologic agents. J Manag Care Pharm. 2004;10:318–25.

8. Bhosle MJ, Feldman SR, Camacho FT, Timothy Whitmire J, Nahata MC, Balkrishnan R. Medication adherence and health care costs associated with biologics in Medicaid‐enrolled patients with psoriasis. J. Dermatol. Treat. 2006;17:294–301.

9. Boudreau R, Blackhouse G, Goeree R, Mierzwinski-Urban M. Adalimumab, alefacept, efalizumab, etanercept, and infliximab for severe psoriasis vulgaris in adults: budget impact analysis and review of comparative clinical-and cost-effectiveness. Ottawa: Canadian Agency for Drugs and Technology in Health; 2007. Report No.: 97.

10. Mauskopf J, Samuel M, McBride D, Mallya UG, Feldman SR. Treatment Sequencing After Failure of the First Biologic in Cost-Effectiveness Models of Psoriasis: A Systematic Review of Published Models and Clinical Practice Guidelines. Pharmacoeconomics. 2014;32:395–409.

11. Glanville J, Fleetwood K, Yellowlees A, Kaunelis D, Mensinkai S. Development and testing of search filters to identify economic evaluations in MEDLINE and EMBASE. Ottawa, ON: Canadian Agency for Drugs and Technologies in Health; 2009. Report No.: H0490.

12. Chiou C-F, Hay JW, Wallace JF, Bloom BS, Neumann PJ, Sullivan SD, et al. Development and validation of a grading system for the quality of cost-effectiveness studies. Med. Care. 2003;41:32–44.

13. Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ Economic Evaluation Working Party. BMJ. 1996;313:275–83.

14. Glennie JL, Torrance GW, Baladi JF, Berka C, Hubbard E, Menon D, et al. The revised Canadian guidelines for the economic evaluation of pharmaceuticals. Pharmacoeconomics. 1999;15:459–68.

15. Drummond MF, Richardson WS, O’Brien BJ, Levine M, Heyland D. Users’ Guides to the Medical Literature: XIII. How to Use an Article on Economic Analysis of Clinical Practice A. Are the Results of the Study Valid? JAMA. 1997;277:1552–7.

16. O’Brien BJ, Heyland D, Richardson WS, Levine M, Drummond MF. Users’ Guides to the Medical Literature: XIII. How to Use an Article on Economic Analysis of Clinical Practice B. What Are the Results and Will They Help Me in Caring for My Patients? JAMA. 1997;277:1802–6.

17. Wong WB, Carlson JJ, Thariani R, Veenstra DL. Cost effectiveness of pharmacogenomics. Pharmacoeconomics. 2010;28:1001–13.

18. Ofman JJ, Sullivan SD, Neumann PJ, Chiou CF, Henning JM, Wade SW, et al. Examining the value and quality of health economic analyses: implications of utilizing the QHES. J Manag Care Pharm. 2003;9:53–61.

19. Hankin CS, Bhatia ND, Goldenberg G, Bronstone A, Dunn JD. A comparison of the clinical effectiveness and cost-effectiveness of treatments for moderate to severe psoriasis. Drug Benefit Trends. 2010;22:17–27.

20. Hankin CS, Feldman SR, Szczotka A, Stinger RC, Fish LS, Hankin DL. A Cost Comparison of Treatments of Moderate to Severe Psoriasis. Drug Benefit Trends. 2005;17:200–14.

21. Menter A, Baker T. Cost-efficacy analysis of biological treatments in psoriasis: an 18-month assessment. J. Med. Econ. 2005;8:139–46.

22. Oh P, Gupta A, Einarson T, Maerov P, Shear N. Calcipotriol in the treatment of psoriasis of limited severity: pharmacoeconomic evaluation. J Cutan Med Surg. 1997;2:7–15.

23. Ahn CS, Gustafson CJ, Sandoval LF, Davis SA, Feldman SR. Cost effectiveness of biologic therapies for plaque psoriasis. Am J Clin Dermatol. 2013;14:315–26.

24. Anis AH, Bansback N, Sizto S, Gupta SR, Willian MK, Feldman SR. Economic evaluation of biologic therapies for the treatment of moderate to severe psoriasis in the United States. J Dermatol. Treat. 2011;22:65–74.

25. Bergstrom KG, Arambula K, Kimball AB. Medication formulation affects quality of life: a randomized single-blind study of clobetasol propionate foam 0.05% compared with a combined program of clobetasol cream 0.05% and solution 0.05% for the treatment of psoriasis. Cutis. 2003;72:407–11.

26. Chen S, Shaheen A, Garber A. Cost-effectiveness and cost-benefit analysis of using methotrexate vs Goeckerman therapy for psoriasis: a pilot study. Arch. Dermatol. 1998;134:1602–8.

27. Ellis CN, Reiter KL, Bandekar RR, Fendrick AM. Cost-effectiveness comparison of therapy for psoriasis with a methotrexate-based regimen versus a rotation regimen of modified cyclosporine and methotrexate. J Am Acad Dermatol. 2002;46:242–50.

28. Feldman SR, Garton R, Averett W, Balkrishnan R, Vallee J. Strategy to manage the treatment of severe psoriasis: considerations of efficacy, safety and cost. Expert Opin Pharmacother. 2003;4:1525–33.

29. Liu Y, Wu EQ, Bensimon AG, Fan CP, Bao Y, Ganguli A, et al. Cost per responder associated with biologic therapies for Crohn’s disease, psoriasis, and rheumatoid arthritis. Adv Ther. 2012;29:620–34.

30. Marchetti A, Feldman SR, Kimball AB, Anderson RR, Miller LH, Martin J, et al. Treatments for mild-to-moderate recalcitrant plaque psoriasis: expected clinical and economic outcomes for first-line and second-line care. Dermatol Online J. 2005;11:1.

31. Marchetti A, LaPensee K, An P. A pharmacoeconomic analysis of topical therapies for patients with mild-to-moderate stable plaque psoriasis: a US study. Clin Ther. 1998;20:851–69.

32. Martin S, Feldman SR, Augustin M, Szapary P, Schenkel B. Cost per responder analysis of ustekinumab and etanercept for moderate to severe plaque psoriasis. J Dermatol. Treat. 2011;22:138–43.

33. Nelson AA, Pearce DJ, Fleischer AB, Balkrishnan R, Feldman SR. New treatments for psoriasis: which biologic is best? J Dermatol. Treat. 2006;17:96–107.

34. Nelson AA, Pearce DJ, Fleischer AB Jr, Balkrishnan R, Feldman SR. Cost-effectiveness of biologic treatments for psoriasis based on subjective and objective efficacy measures assessed over a 12-week treatment period. J Am Acad Dermatol. 2008;58:125–35.

35. Pan F, Brazier NC, Shear NH, Jivraj F, Schenkel B, Brown R. Cost utility analysis based on a head-to-head Phase 3 trial comparing ustekinumab and etanercept in patients with moderate-to- severe plaque psoriasis: a Canadian perspective. Value Health. 2011;14:652–6.

36. Pearce DJ, Nelson AA, Fleischer AB, Balkrishnan R, Feldman SR. The cost-effectiveness and cost of treatment failures associated with systemic psoriasis therapies. J Dermatol. Treat. 2006;17:29–37.

37. Poulin Y, Langley RG, Teixeira HD, Martel MJ, Cheung S. Biologics in the treatment of psoriasis: clinical and economic overview. J Cutan Med Surg. 2009;13 Suppl 2:S49–57.

38. Staidle JP, Dabade TS, Feldman SR. A pharmacoeconomic analysis of severe psoriasis therapy: a review of treatment choices and cost efficiency. Expert Opin Pharmacother. 2011;12:2041–54.

39. Stern RS. The benefits, costs and risks of topical tar preparations in the treatment of psoriasis: considerations of cost effectiveness. Ann Acad Med Singap. 1988;17:473–6.

40. Villacorta R, Hay JW, Messali A. Cost effectiveness of moderate to severe psoriasis therapy with etanercept and ustekinumab in the United States. Pharmacoeconomics. 2013;31:823–39.

41. Blasco AJ, Lazaro P, Ferrandiz C, Garcia-Diez A, Liso J. Efficiency of biologic agents in the treatment of moderate to severe psoriasis. Actas Dermosifiliogr. 2009;100:792–803.

42. De Argila D, Rodriguez-Nevado I, Chaves A. Cost-effectiveness analysis comparing methotrexate with PUVA therapy for moderate-severe psoriasis in the sanitary area of Badajoz. Actas Dermosifiliogr. 2007;98:35–41.

43. Colombo GL, Di Matteo S, Peris K, Fargnoli MC, Esposito M, Mazzotta A, et al. A cost- utility analysis of etanercept for the treatment of moderate-to-severe psoriasis in Italy. Clin. Outcomes Res. 2009;1:53–9.

44. Colombo GL, Di Matteo S, Bruno G, Girolomoni G, Vena GA. Calcipotriol and betamethasone dipropionate in the treatment of mild-to-moderate psoriasis: a cost-effectiveness analysis of the ointment versus gel formulation. Clin. Outcomes Res. 2012;4:261–8.

45. Harrington C. Cost-effectiveness analysis of calcipotriol ointment and“ short-contact” dithranol in treating mild-to-moderate psoriasis. Br. J. Med. Econ. 1995;8:27–32.

46. De Portu S, Del Giglio M, Altomare G, Arcangeli F, Berardesca E, Pinton PC, et al. Cost- effectiveness analysis of TNF-α blockers for the treatment of chronic plaque psoriasis in the perspective of the Italian health-care system. Dermatol. Ther. 2010;23:S7–13.

47. Vano-Galvan S, Garate MT, Fleta-Asin B, Hidalgo A, Fernandez-Guarino M, Bermejo T, et al. Analysis of the cost effectiveness of home-based phototherapy with narrow-band UV-B radiation compared with biological drugs for the treatment of moderate to severe psoriasis. Actas Dermosifiliogr. 2012;103:127–37.

48. Ashcroft DM, Li Wan Po A, Williams HC, Griffiths CE. Cost-effectiveness analysis of topical calcipotriol versus short-contact dithranol. In the treatment of mild to moderate plaque psoriasis. Pharmacoeconomics. 2000;18:469–76.

49. Augustin M, Peeters P, Radtke M, Moehling U, Lapp C. Cost-effectiveness model of topical treatment of mild to moderate psoriasis vulgaris in Germany. A comparison of calcipotriol/betamethasone (Daivobet/Dovobet/Taclonex) once daily and a morning/evening non-fix combination of calcipotriol and betamethasone. Dermatology. 2007;215:219–28.

50. Augustin M, Radtke M, van Engen A, Ruedig C, Lapp C, Moehling U. Pharmacoeconomic model of topical treatment options of mild to moderate psoriasis vulgaris in Germany. J Dtsch Dermatol Ges. 2009;7:329–38.

51. Bottomley JM, Auland ME, Morais J, Boyd G, Stewart Douglas W. Cost-effectiveness of the two-compound formulation calcipotriol and betamethasone dipropionate compared with commonly used topical treatments in the management of moderately severe plaque psoriasis in Scotland. Curr. Med. Res. Opin. 2007;23:1887–901.

52. Ferrandiz C, Garcia A, Blasco AJ, Lazaro P. Cost-efficacy of adalimumab, etanercept, infliximab and ustekinumab for moderate-to-severe plaque psoriasis. J Eur Acad Dermatol Venereol. 2012;26:768–77.

53. Freeman K, Marum M, Bottomley JM, Auland M, Jackson P, Ryttov J. A psoriasis-specific model to support decision making in practice - UK experience. Curr Med Res Opin. 2011;27:205–23.

54. Greiner RA, Braathen LR. Cost-effectiveness of biologics for moderate-to-severe psoriasis from the perspective of the Swiss healthcare system. Eur J Dermatol. 2009;19:494–9.

55. Hartman M, Prins M, Swinkels OQ, Severens JL, De Boo T, Van Der Wilt GJ, et al. Cost- effectiveness analysis of a psoriasis care instruction programme with dithranol compared with UVB phototherapy and inpatient dithranol treatment. Br J Dermatol. 2002;147:538–44.

56. Heinen-Kammerer T, Daniel D, Stratmann L, Rychlik R, Boehncke WH. Cost-effectiveness of psoriasis therapy with etanercept in Germany. J Dtsch Dermatol Ges. 2007;5:762–8.

57. Knight C, Mauskopf J, Ekelund M, Singh A, Yang S, Boggs R. Cost-effectiveness of treatment with etanercept for psoriasis in Sweden. Eur J Health Econ. 2012;13:145–56.

58. Koek MB, Sigurdsson V, van Weelden H, Steegmans PH, Bruijnzeel-Koomen CA, Buskens E. Cost effectiveness of home ultraviolet B phototherapy for psoriasis: economic evaluation of a randomised controlled trial (PLUTO study). BMJ. 2010;340:c1490.

59. Lloyd A, Reeves P, Conway P, Reynolds A, Baxter G. Economic evaluation of etanercept in the management of chronic plaque psoriasis. Br J Dermatol. 2009;160:380–6.

60. Sawyer L, Samarasekera EJ, Wonderling D, Smith CH. Topical therapies for the treatment of localized plaque psoriasis in primary care: a cost-effectiveness analysis. Br J Dermatol. 2013;168:1095–105.

61. Schmitt-Rau K, Rosenbach T, Radtke MA, Augustin M. Cost-effectiveness of biological therapy in remission induction of moderate to severe plaque psoriasis. Dermatology. 2010;221:236–42.

62. Sizto S, Bansback N, Feldman SR, Willian MK, Anis AH. Economic evaluation of systemic therapies for moderate to severe psoriasis. Br J Dermatol. 2009;160:1264–72.

63. Snellman E, Maljanen T, Aromaa A, Reunanen A, Jyrkinen-Pakkasvirta T, Luoma J. Effect of heliotherapy on the cost of psoriasis. Br J Dermatol. 1998;138:288–92.

64. Papp K, Poulin Y, Barber K, Lynde C, Prinz JC, Berg M, et al. Cost-effectiveness evaluation of clobetasol propionate shampoo (CPS) maintenance in patients with moderate scalp psoriasis: a Pan-European analysis. J Eur Acad Dermatol Venereol. 2012;26:1407–14.

65. Peeters P, Ortonne JP, Sitbon R, Guignard E. Cost-effectiveness of once-daily treatment with calcipotriol/betamethasone dipropionate followed by calcipotriol alone compared with tacalcitol in the treatment of Psoriasis vulgaris. Dermatology. 2005;211:139–45.

66. Devaux S, Castela A, Archier E, Gallini A, Joly P, Misery L, et al. Topical vitamin D analogues alone or in association with topical steroids for psoriasis: a systematic review. J. Eur. Acad. Dermatol. Venereol. 2012;26:52–60.

67. Woolacott N, Hawkins N, Mason A, Kainth A, Khadjesari Z, Vergel YB, et al. Etanercept and efalizumab for the treatment of psoriasis: a systematic review. Health Technol. Assess. 2006;10.

68. Hakkaart-van Roijen L, Verboom P, Redekop WK, Touw KR, Rutten FF. The cost effectiveness of tapered versus abrupt discontinuation of oral cyclosporin microemulsion for the treatment of psoriasis. Pharmacoeconomics. 2001;19:599–608.

69. Aggarwal K, Khandpur S, Khanna N, Sharma VK, Pandav CS. Comparison of clinical and cost-effectiveness of psoralen + ultraviolet A versus psoralen + sunlight in the treatment of chronic plaque psoriasis in a developing economy. Int J Dermatol. 2013;52:478–85.

70. Shani J, Harari M, Hristakieva E, Seidl V, Bar-Giyora J. Dead-Sea climatotherapy versus other modalities of treatment for psoriasis: comparative cost-effectiveness. Int J Dermatol. 1999;38:252–62.

71. Igarashi A, Kuwabara H, Fahrbach K, Schenkel B. Cost-efficacy comparison of biological therapies for patients with moderate to severe psoriasis in Japan. J Dermatol. Treat. 2013;24:351–5.

72. Kerkhof PCM. On the limitations of the psoriasis area and severity index (PASI). Br. J. Dermatol. 1992;126:205–6.

73. Jacobson CC, Kimball AB. Rethinking the Psoriasis Area and Severity Index: the impact of area should be increased. Br. J. Dermatol. 2004;151:381–7.

74. Feldman SR, Fleischer AB, Reboussin DM, Rapp SR, Exum ML, Clark AR, et al. The self- administered psoriasis area and severity index is valid and reliable. J. Invest. Dermatol. 1996;106:183–6.

75. Harari M, Shani J, Hristakieva E, Stanimirovic A, Seidl W, Burdo A. Clinical evaluation of a more rapid and sensitive Psoriasis Assessment Severity Score (PASS), and its comparison with the classic method of Psoriasis Area and Severity Index (PASI), before and after climatotherapy at the Dead‐Sea. Int. J. Dermatol. 2000;39:913–8.

76. Menter MA, Krueger GC, Feldman SR, Weinstein GD. Psoriasis treatment 2003 at the new millennium: position paper on behalf of the authors. J. Am. Acad. Dermatol. 2003;49:39–43.

77. Carlin CS, Feldman SR, Krueger JG, Menter A, Krueger GG. A 50% reduction in the Psoriasis Area and Severity Index (PASI 50) is a clinically significant endpoint in the assessment of psoriasis. J. Am. Acad. Dermatol. 2004;50:859–66.

78. Räsänen P, Roine E, Sintonen H, Semberg-Konttinen V, Ryynänen O-P, Roine R. Use of quality-adjusted life years for the estimation of effectiveness of health care: A systematic literature review. Int. J. Technol. Assess. Health Care. 2006;22:235–41.

79. Schmitt JM, Ford DE. Work limitations and productivity loss are associated with health- related quality of life but not with clinical severity in patients with psoriasis. Dermatol. Basel Switz. 2006;213:102–10.

80. Pearce DJ, Singh S, Balkrishnan R, Kulkarni A, Fleischer AB, Feldman SR. The negative impact of psoriasis on the workplace. J. Dermatol. Treat. 2006;17:24–8.

81. All Wales Medicines Strategy Group. Final Appraisal Report: Etanercept (Enbrel®). All Wales Medicines Strategy Group; 2010.

82. National Institute for Health and Clinical Excellence. Ustekinumab for the treatment of adults with moderate to severe psoriasis. London: National Institute for Health and Clinical Excellence (NICE); 2009. Report No.: 180.

83. National Institute for Health and Clinical Excellence. Infliximab for the treatment of adults with psoriasis. London: National Institute for Health and Clinical Excellence (NICE); 2008. Report No.: 134.

Articles identified through Additional articles database* search identified by hand search (n=2,617) (n=10)

Duplicate articles (n=710)

Titles/abstracts screened after duplicates were excluded (n=1,917) Stage 1 Excluded (n=1,417)  No abstract (n=8)  Not about Psoriasis/Psoriasis Treatment (n=868)  Not Economic Evaluation (n=452)  Others* (n=89) *Others include: conference abstracts, letters/editorials, conference reports, errata, and book chapters.

Full text articles screened for eligibility (n=500)

Stage 2 Excluded (n=447)  Burden of Illness & Cost Analysis (n=98)  Not Economic Evaluation (n=240)  Review (n=39)  Editorials/Reports (n=20)  Non-English (n=48)  Not downloadable (n=2)

Articles included in the final review (n=53)

* NHS EED (60); HTA (60); CEA Registry (15); Medline (695); Embase (1,787)

Figure 1: PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) flowchart of inclusion and exclusion of articles in the review.

Table 1: Proportion of articles reporting each of the items* of Quality of Health Economic Studies instrument

Item Description N (%) 1. Was the study objective presented in a clear, specific, and measurable manner? 52 (98.1) 2. Were the perspective of the analysis (societal, third-party payer, etc.) and reasons 39 (73.6) for its selection stated? 3. Were variable estimates used in the analysis from the best available source (i.e. 52 (98.1) Randomized Control Trial-Best, Expert Opinion-Worst)? 4. If estimates came from a subgroup analysis, were the groups pre-specified at the NA beginning of the study? 5. Was uncertainty handled by: 1) statistical analysis to address random events; 2) 38 (71.7) sensitivity analysis to cover a range of assumptions? 6. Was incremental analysis performed between alternatives of resources and costs? 33 (62.3) 7. Was the methodology for data abstraction (including value health states and other 53 (100.0) benefits) stated? 8. Did the analytic horizon allow time for all relevant and important outcomes? Were benefits and costs that went beyond 1 year discounted (3-5%) and 22 (41.5) justification given for the discount rate? 9. Was the measurement of costs appropriate and the methodology for the estimation 33 (62.3) of quantities and unit costs clearly described? 10. Were the primary outcome measure(s) for the economic evaluation clearly stated 52 (98.1) and were the major short term, long term and negative outcomes included? 11. Were the health outcomes measures/scales valid and reliable? If previously tested valid and reliable measures were not available, was justification given for the 49 (92.5) measures/scales used? 12. Were the economic model (including structure) study methods and analysis, and the components of the numerator and denominator displayed in a clear transparent 44 (83.0) manner? 13. Were the choice of economic model, main assumptions and limitations of the 48 (90.6) study stated and justified? 14. Did the author(s) explicitly discuss direction and magnitude of potential biases? 13 (24.5)

15. Were the conclusions/recommendations of the study justified and based on the 53 (100.0) study results? 16. Was there a statement disclosing the source of funding for the study? 36 (67.9) NA=Not applicable (in all but 5 of the articles reviewed).

Table 2: Summary of Economic Analyses

Short-contact Dithranol as Degree of first line treatment may Ashcroft et al., Cost-effectiveness of Mild-to- UK National 12 weeks up to improvement in help contain costs and 2000; UK; 2000; topical calcipotriol and Calcipotriol vs. Dithranol moderate Drug costs Health Service 1 year psoriasis as judged by improve outcomes in terms £ [48] short-contact dithranol psoriasis the patient of more durable remission following treatment DCD (disease controlled days); Cost-effectiveness of fix patient's assessment Fix Fix Calcipotriol/Betamethasone Calcipotriol/Betamethasone of change from Medication unit costs; Calcipotriol/Betamethason Augustin et al., combination vs. a German Mild-to- combination compared to a baseline following a adverse events; UVB e combination is more 2007; Germany; morning/evening non-fix Societal moderate 48 weeks morning/evening non-fix 6-point scale rescue therapy per cost-effective than a non- 2006(?); € [49] Calcipotriol/Betamethasone Perspectives psoriasis Calcipotriol/Betamethasone (clearance, marked, session and visit fix morning/evening combination combination moderate or slight combination improvement, no change, worsening) A compound product containing Calcipotriol/Betamethasone, given once daily for 4 weeks, Fix followed by daily Calcipotriol Calcipotriol/Betamethason Evaluate and compare Drug costs; costs Augustin et al., for 4 more weeks; (2) Mild-to- e combination is a more effectiveness and cost of Not DCD (disease related to adverse 2009; Germany; Tacalcitol, given once daily for moderate 48 weeks cost-effective treatment various topical psoriasis mentioned controlled days) effects; UVB 2007(?); € [50] 8 weeks; (3) Separate psoriasis than a treatment with the therapies phototherapy administration in the single agents or Tacalcitol morning/evening of Calcipotriol monotherapy and Betamethasone, twice daily, for 8 weeks To test whether or not No significant difference different formulations of the Clobetasol foam to affected skin in cost was appreciated Bergstrom et al., same medication—topical and scalp vs. clobetasol cream Not Moderate-to- 1-point change in between foam and 2003; USA; NR; 2 weeks Drug costs clobetasol propionate— to the skin and clobetasol mentioned severe psoriasis PASI score cream/solution over the USD [25] would affect measurable solution to the scalp period after controlling for outcomes body surface area With reduced costs and Drug costs; GP superior outcomes, the Cost-effectiveness of consultation; specialist TCF (two compound Bottomley et al., Calcipotriol/Betamethasone National outpatient consultant Moderate-to- formulation) ‘dominated’ 2007; UK; 2006- dipropionate in the initial Multiple treatment sequence* Health Service 1 year QALY consultation; specialist severe psoriasis the other treatments since 2007; £ [51] treatment of moderate in Scotland outpatient nurse the latter were associated severity plaque psoriasis consultation; nurse-led with higher cost and lower phototherapy course utility or QALY gain Italian The gel strategy appears to Cost-effectiveness of a gel Colombo et al., Calcipotriol and Betamethasone National Mild-to- Costs of medication; be favorable from the containing calcipotriol and 2012; Italy; dipropionate vs. the ointment Healthcare moderate 1 year PASI-75 fees for specialist and pharmacoeconomic point betamethasone dipropionate 2012; € [44] formulation System as a psoriasis GP visits of view than the ointment vs. the ointment formulation third-party formulation

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency payer Cost-efficacy of vitamin D PASI-90 (Treatment The cost ⁄ efficacy ratio Devaux et al., Vitamin D analogues plus analogues plus topical Not success), and PASI- was evaluated as 1.2–1.8 2012; France; topical steroids (VDS) vs. Not mentioned 4 weeks Drug costs steroids (VDS) vs. vitamin mentioned 75 (Satisfactory times higher for VDS than 2010; € [66] vitamin D analogues alone (VD) D analogues alone (VD) response) for VD Calcipotriol should be Harrington, Cost-effectiveness of Calcipotriol ointment twice Mild-to- used in any national 1995; UK; 1994 calcipotriol ointment and Not Own outcome criteria daily for 8 weeks vs. dithranol moderate 8 weeks Drug costs treatment program for (NHS cost); "short-contact" dithranol mentioned from a scale of 1 to 4† once daily psoriasis mild-to-moderate psoriasis £ [45] regimen over dithranol Cost-effectiveness analysis Costs for physician Marchetti et al., of topical therapies for Tazarotene 0.1%, Tazarotene Mild-to- visits, drug acquisition, Third-party Tazarotene 0.1% was the 1998; USA; patients with mild-to- 0.05%, Fluocinonide, and moderate 1 year Disease-Free Days laboratory testing, and payer most cost-effective option 1997; USD [31] moderate stable plaque Calcipotriene psoriasis adverse events psoriasis management Costs of physician visits, laboratory tests, UVB therapy; the cost Determine cost- of PUVA therapy Calcipotriene is cost- effectiveness of calcipotriol (including costs of effective alternative to compared with medium- to psoralen tablets, Calcipotriene (CP), medium- to high-potency Oh et al., 1997; high-potency steroids in the Government Mild-to- physician fees, Betamethasone dipropionate corticosteroids, both as Canada; 1995; management of psoriasis of payer moderate 1 year QALY laboratory tests, and (BD), Betamethasone valerate second-line therapy to CAD [22] limited extent that had perspective psoriasis facilities fees to (BV), Fluocinonide (F) Betamethasone valerate previously been treated with provide PUVA (BV) or when failure is betamethasone-17-valerate, therapy;’; costs with BV 0.1% associated with failures and relapses (i.e., additional treatments, visits, and tests) Bemethson ointment Betnovat Papp et al., emulsion ⁄ solution, Diprolene 2012; Belgium, cream, Celestan V ointment, Denmark, Clarelux foam, Ecural solution Finland, France, Cost-effectiveness Elucon solution + bucky Elucon Germany, evaluation of Clobetasol Clobetasol propionate solution, Diprosalic lotion ⁄ Cost of physician visits Norway, Propionate Shampoo (CPS) Payer Moderate-to- shampoo (CPS) is cost- Psodermil solution, Elocom 24 weeks Disease-Free Days and cost of Portugal, maintenance in patients perspective severe psoriasis effective in maintaining solution, Dermovate sol ⁄ interventions Sweden, and with moderate scalp the success achieved Dermoval sol, Daivonex Switzerland; psoriasis solution, Dovobet-Daivobet NR; Respective lotion ⁄ Xamiol gel, Vehicle, No local currency further treatment, Maintenance [64] CPS, Acute CPS Cost-effectiveness of once- daily treatment with Peeters et al., Drug costs, hospital Calcipotriol/Betamethasone Mean PASI 2005; France, French stays, days of hospital Calcipotriol/Betamethason Dipropionate followed by Daivobet/Daivonex, and score ≈10; Germany, Spain societal 8 weeks PASI-75 attendance, physician e is more effective and less Calcipotriol alone compared Tacalcitol Mean age≈51 and UK; 2004; € perspective visits, lab tests, and costly than Tacalcitol with Tacalcitol in the years [65] costs of adverse events treatment of psoriasis vulgaris

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency Potent corticosteroids, used alone or in combination with vitamin D, are the most cost- Costs of topical agents, Cost-effectiveness and Patients with effective treatment for Sawyer et al., UK National 1 year; primary and secondary optimal treatment sequence Various combination of psoriasis of the patients with psoriasis of 2013; UK; 2011; Health Service extended to 3 QALY care visits and second- for psoriasis of the trunk, treatment sequence trunk, limbs the trunk and limbs. Potent £ [60] perspective and 10 years line therapies for limbs and scalp and scalp or very potent treatment failures corticosteroids are the most cost-effective treatment for patients with scalp psoriasis. PHOTOTHERAPY

Costs of drugs, payment for phototherapy; Compare efficacy and cost- consultation fee; PUVAsol had a clinical Aggarwal et al., Patient's and effectiveness of PUVA vs. Moderate-to- Percentage of transportation or travel efficacy comparable with 2013; India; NR; PUVA vs. PUVAsol hospital's 12 weeks PUVAsol in chronic plaque severe psoriasis improvement in PASI cost; wages or salary PUVA and favourable USD [69] perspective psoriasis lost; equipment cost; cost-effectiveness ratio overhead costs; salaries of doctors and other staff Outpatient phototherapy; consultation with 17.6 weeks dermatologist, and GP; (mean duration medication travelling Costs and cost-effectiveness at the end of costs; parking costs for Home UVB phototherapy Koek et al., of phototherapy with UVB phototherapy); visits to hospital; is not more expensive than 2010; Home vs. outpatient UVB Societal Mild-to-severe light provided at home 68.4 weeks QALY parking costs for visits phototherapy in an Netherlands; phototherapy perspective psoriasis compared with outpatient (mean duration to GP; absence from outpatient setting and 2003; € [58] ultraviolet B phototherapy one year after paid work; reduced proved to be cost effective the end of productivity while at phototherapy paid work; absence from unpaid work; side effects (that did not vary across groups) Ward treatment, The cost of heliotherapy phototherapy in exceeded manifold the outpatient clinics, lab mean monthly cost of X-ray, physician conventional psoriasis Snellman et al., consultations, therapy. There was no 1998; Spain 1 year before Difference in mean Cost-effectiveness of Time; before and after Not Moderate-to- medication at home, overall savings using (Finnish and after a 4- Psoriasis Severity heliotherapy Heliotherapy mentioned severe psoriasis trips for treatments, heliotherapy in moderate Patients); 1988- week treatment Index (PSI) ‡ self-arranged sun- to moderately severe 90; FIM [63] bathing holidays, psoriasis; it saved costs productivity loss due to only in severe psoriasis absenteeism and that required expensive unpaid help medication. SYSTEMICS

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency TRADITIONAL SYSTEMICS Cost-effectiveness for treating psoriasis using two strategies: one consisted Acquisition of Ellis et al., principally of methotrexate Methotrexate-based vs. medications, laboratory $2700 per incremental year Payers’ Moderate-to- Years clear of 2002; USA; and the other was cyclosporine-methotrexate 10 years and physician fees, and clear of psoriasis with the perspective severe psoriasis psoriasis 1999; USD [27] principally a rotational rotation treatment strategies costs of treating side cyclosporine strategy schedule of modified effects cyclosporine with methotrexate. Costs of medication; Patients with visits to the Hakkaart-van The cost-effectiveness of chronic plaque dermatologist; Roijen et al., tapered versus abrupt Tapering cyclosporine was Tapered versus abrupt psoriasis Number of systemic laboratory tests; costs 2001; Canada, discontinuation of oral Societal more cost effective than discontinuation of oral inadequately 1 year therapy-free days of lost production due Spain, Turkey, cyclosporine micro- perspective abruptly stopping cyclosporine controlled with (STFDs) to illness and/or its UK; 1997; USD emulsion for the treatment cyclosporine topical treatment costs; [68] of psoriasis treatment adverse events (did not vary across groups) BIOLOGICS

Intravenous infliximab 3 Determine and compare the mg/kg was the most cost- 12 weeks; cost-effectiveness of Drug costs; physician effective biologic agent Ahn et al., 2013; Adalimumab, Alefacept, modeled for biologics with regard to cost Third-party Moderate-to- visits; lab tests; with respect to both the USA; 2010; Etanercept, Infliximab, annual cost- PASI and DLQI MID per patient achieving DLQI payer severe psoriasis intravenous infusion cost per patient achieving USD [23] Ustekinumab; placebo effectiveness MID, and cost per patient procedures PASI-75 and the cost per analysis as well achieving PASI-75 patient achieving a DLQI MID Drug and related costs; Optimal sequence in Patients who hospitalization costs; Anis et al., Economic evaluation of Adalimumab, Etanercept, Not clear; prescribing biologics will Not failed productivity cost; 2011; USA; biologics for moderate to Infliximab, Alefacept, annualized QALY be: Adalimumab mentioned conventional justified for not using 2007; USD [24] severe psoriasis Efalizumab results reported Etanercept Infliximab therapies the cost of adverse Efalizumab Alefacept events Estimate the cost/efficacy Spanish ratios of Adalimumab, The most efficient biologic Blasco et al., National Etanercept, Infliximab, and Adalimumab, Infliximab, Moderate-to- agent in terms of the 2009; Spain; Health System 10-24 weeks PASI-75 Drug cost only Efalizumab in the Etanercept, Efalizumab severe psoriasis cost/efficacy ratio was 2008; € [41] payer management of moderate to Adalimumab perspective severe psoriasis (1) After 24 weeks: Infliximab (5 mg/kg), Etanercept (25 or 50 Cost-effectiveness analysis mg biw, 50 mg weekly, step Italian health- 24 weeks, and de Portu et al., of Infliximab compared down), Adalimumab (40 eow); Direct medical costs Infliximab seems to be care system Moderate-to- 48-50 weeks 2010; Italy; with other anti-TNF-α (2) After 48-50 weeks: PASI-75 with specific reference cost-effective in the (third party severe psoriasis (based on 2008; € [46] agents for the treatment of Infliximab (5 mg/kg), to the cost of therapy therapy of psoriasis payer) RCTs) psoriasis in Italy Etanercept (50 mg biw, step down), Adalimumab (40 mg eow) Ferrandiz et al., Cost-efficacy of Adalimumab, Etanercept, Spanish Moderate-to- Adalimumab was the most 24 weeks PASI-75 Only drug cost 2012; Spain; Adalimumab, Etanercept, Infliximab and Ustekinumab National severe psoriasis cost-efficient

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency 2010; € [52] Infliximab and Health System Ustekinumab for moderate- payer to-severe plaque psoriasis perspective Drug acquisition costs Greiner & (i.e. medication, Cost-effectiveness of Infliximab, Etanercept, Swiss Infliximab and Braathen, 2009; Moderate-to- PASI-50, PASI-75, administration, and biologics for moderate-to- Adalimumab, Efalizumab, healthcare 36 weeks Adalimumab generated the Switzerland; severe psoriasis and PASI-90 monitoring) and costs severe psoriasis Alefacept system lowest ICERs at 36 weeks 2006; CHF [54] incurred by adverse events Japanese Igarashi et al., Cost-efficacy comparison of National Ustekinumab was more Adalimumab, Infliximab, Moderate-to- 2013; Japan; biologics for patients with Health 2 years PASI-75 Drug costs cost-efficient than Ustekinumab severe psoriasis NR; USD [71] moderate to severe psoriasis Insurance Adalimumab or Infliximab (NHI) Adalimumab 40 mg eow, Liu et al., 2012; Cost per responder Etanercept 50 mg BIW, 3-month cost per responder Not Moderate-to- PASI-75 and PASI- Drug acquisition and USA; 2011; associated with biologic Ustekinumab 45 mg, 52 weeks was lowest for mentioned severe psoriasis 90 administration costs USD [29] therapies Ustekinumab 90 mg, Infliximab Adalimumab 5 mg/kg Martin et al., Cost-effectiveness analysis The cost per responder was Not Moderate-to- 2011; USA; NR; of Ustekinumab and Ustekinumab and Etanercept 16 weeks PASI-75 Drug costs lower for Ustekinumab mentioned severe psoriasis USD [32] Etanercept than for Etanercept 1) Alefacept 15 mg IM weekly for two 12-week courses; Drug costs, IM/SC 2) Efalizumab 1 mg/kg SC injection fees, office Menter & Baker, Cost-efficacy of Alefacept, Cost-efficacy of Alefacept, weekly; Managed care Moderate-to- visit fees, costs for 2005; USA; Efalizumab, or Etanercept 18 months PASI-75 Efalizumab and Etanercept 3) Etanercept 50 mg SC twice perspective severe psoriasis laboratory monitoring; 2003; USD [21] for plaque psoriasis was comparable weekly for 12 weeks followed costs incurred due to by a maintenance dose of 50 mg adverse events weekly Drug cost; costs of the Adalimumab and Nelson et al., Cost-effectiveness of Alefacept, Efalizumab, Not Severe PASI-75 and Mean physician visits; Infliximab appear to be the 2006; USA; biologics in the treatment of Etanercept, Infliximab, and 12 weeks mentioned psoriasis DLQI improvement required laboratory most cost-effective 2004; USD [33] psoriasis Adalimumab testing, and infusions biologic agents Etanercept at a dose of 25 mg administered Mean age subcutaneously once range: 43-47 Costs of drug, lab, weekly was the most cost- years; physician, and effective agent in cost per Nelson et al., Cost-effectiveness of Adalimumab, Alefacept, Third-party predominantly PASI-75 and DLQI infusion; justification patient achieving DLQI 2008; USA; biologic treatments for Efalizumab, Etanercept, 12 weeks payer male; mean MID given for not using MID; Infliximab IV at a 2006; USD [34] psoriasis Infliximab baseline PASI long-term adverse dose of 3 mg/kg was the Score: 14.2- effects most cost-effective agent 23.4 in terms of cost per patient achieving PASI-75 improvement Cost-effectiveness of Ustekinumab dominated Pan et al., 2011; Ustekinumab and Ontario 12 weeks Moderate-to- Drug cost; monitoring Etanercept because of Canada; NR; Etanercept in patients with Ustekinumab vs. Etanercept Ministry of (extended to 10 QALY severe psoriasis cost; outpatient visits lower costs and higher CAD [35] moderate-to-severe plaque Health years) utility values psoriasis

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency Poulin et al., Cost-effectiveness of Adalimumab, Ustekinumab, Not Moderate-to- Adalimumab is the most 2009; Canada; biologics in the treatment of Infliximab, Etanercept, 1 year PASI-75 Drug costs mentioned severe psoriasis cost-effective 2009; CAD [37] psoriasis Efalizumab, Alefacept Infliximab at a dose of 3 Cost-effectiveness of Drug costs; cost of mg/kg was the most cost- Schmitt-Rau et biological therapy in German third- physician visits; effective agent. However, al., 2010; Adalimumab, Etanercept, Moderate-to- remission induction of party payer’s 12 weeks PASI-75 laboratory and marked overlap of cost- Germany; 2009; Infliximab, and Ustekinumab severe psoriasis moderate to severe plaque perspective monitoring costs, and effectiveness ratios was € [61] psoriasis costs of chest X-rays observed in sensitivity analysis Direct cost; physician visit; costs for receiving subcutaneous Ustekinumab 45 mg is the Villacorta et al., Cost-effectiveness of 12 weeks injections at the US societal Moderate-to- most cost-effective 2013; USA; Ustekinumab compared Ustekinumab vs. Etanercept (extended to 3 QALY physician’s office; perspective severe psoriasis compared to Etanercept 50 2011; USD [40] with Etanercept years) traveling, waiting, and mg therapy actually receiving treatment at the physician’s office Cost of drugs and of their administration Both Etanercept and and monitoring, and Efalizumab could be cost- Woolacott et al., Cost-effectiveness of Etanercept, Efalizumab, and Not Moderate-to- the cost of outpatient effective depending on 2006; UK; 2004- 10 years QALY Etanercept and Efalizumab Supportive Care mentioned severe psoriasis visits and of inpatient patient characteristics and 05; £ [67] stays; justification the threshold the NHS is given for not using cost willing to pay per QALY of adverse events MIXED

Mixed findings: in severe psoriasis, only methotrexate demonstrates a net benefit. Both therapies were cost- effective compared with no Cost-effectiveness of Clearing rates; Utility Costs for supplies, therapy. Liquid Chen et al., Mild-to- Methotrexate vs. Methotrexate vs. Goeckerman US societal assessed by laboratory tests, and methotrexate should be 1998; USA; NR; moderate 52 weeks Goeckerman therapy for therapy perspective willingness-to-pay or medications; physician chosen over the tablet form USD [26] psoriasis Psoriasis visual analog scale fees and hospital fees since it was cheaper and had the same outcome. Goeckerman was cost- effective against liquid methotrexate in severe, but not in mild or moderate psoriasis Hospitalization; day- Italian hospital admissions; Cost-effectiveness of Intermittent Etanercept is a Colombo et al., National specialist medical intermittent therapy with Etanercept vs. non-systemic Moderate-to- cost-effective therapeutic 2009; Italy; Health 10 years QALY examinations; Etanercept in comparison therapy severe psoriasis option compared with non- 2008; € [43] Service’s laboratory tests and with non-systemic therapy systemic therapy perspective instrumental investigations; costs of

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency phototherapy or drug therapies Societal PUVA more cost-effective (direct + Drug costs; physician than with methotrexate. Cost-effectiveness analysis indirect costs) visits; costs of PUVA de Argila et al., However, indirect costs comparing methotrexate to and Public sessions; follow-up 2007; Spain; Moderate-to- (borne by patients in the PUVA for moderate to Methotrexate vs. PUVA Health Service 1 year PASI-50 tests; treatment to 2002/2004; € severe psoriasis Spanish Health System), severe chronic plaque of adverse reactions; costs [42] are higher for PUVA psoriasis Extremadura of transport and lost therapy, a fact that raises (direct costs working time an issue of equity only) Methotrexate, UVB, PUVA, Feldman et al., Drug costs; costs of Cost-effectiveness of Acitretin, Cyclosporine, Third-party Moderate-to- Methotrexate was the most 2003; USA; 1 year PASI-75 office visits during psoriasis therapy Etanercept, Infliximab, payer severe psoriasis cost-effective 2002; USD [28] treatment; lab works Alefacept Calcipotriol b.d. as first line and Steroid as second Freeman et al., line, AND Calcipotriol b.d. Cost-effectiveness of Topical, phototherapy, systemic Primary care Moderate-to- Drug costs; GP 2011; UK; 2008- 2 years QALY as first line and Two- psoriasis therapy in UK and biologics perspective severe psoriasis consultation fees 09; £ [53] compound formulation (TCF) are both cost- effective Costs for medication or phototherapy; Oral systemic medications, treatment UV therapy, and UV Hankin et al., Cost-effectiveness of US managed administration; Biologics vs. oral systemic Moderate-to- PASI-50 and PASI- therapy combined with 2005; USA; phototherapy, oral systemic health care 1 year monitoring for medications vs. phototherapy severe psoriasis 75 Acitretin appear to be the 2004; USD [20] agents, and biologics systems potential treatment- most cost-effective related adverse events, therapies and treatment of adverse events Drug wholesale acquisition cost; administration of IV Hankin et al., Cost-effectiveness of US managed infusion or Biologics vs. oral systemic Moderate-to- Methotrexate was the most 2010; USA; treatments for moderate to health care 1 year PASI-75 phototherapy; medications vs. phototherapy severe psoriasis cost-effective 2008; USD [19] severe psoriasis systems Alefacept IM injection and Infliximab IV infusion (where applicable) Drug costs; hospital stay; use of the UVB unit including time of Cost-effectiveness analysis nurse; use of the day- Hartman et al., of a psoriasis care Psoriasis care instruction care unit including time PASI-90, and the 2002; instruction programme with programme with dithranol vs. Societal Moderate-to- of nurse; dermatologist UVB was most cost- 1 year number of clearance- Netherlands; dithranol compared with UVB phototherapy and inpatient perspective severe psoriasis fee; nurse fee; effective days 1998; € [55] UVB phototherapy and dithranol treatment outpatients visits inpatient dithranol treatment dermatologist fee; GP fee; outpatients visits other specialists fee; absence from paid

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency work; travelling costs Heinen- Cost-effectiveness of Etanercept is a cost- Perspectives Drug costs; physician's Kammerer et al., psoriasis therapy with Etanercept vs. non-systemic Moderate-to- effective measure within of health 10 years QALY fee; costs of side 2007; Germany; Etanercept in comparison to therapy severe psoriasis the German healthcare insurance effects NR; € [56] non-systemic therapy system Drug cost; resource cost (administration as an outpatient visit); cost per initial Once-weekly Etanercept Knight et al., Non-systemic therapy, treatment; cost per re- 50 mg, used intermittently, Cost-effectiveness of Swedish 2012; Sweden; Etanercept 50 mg once weekly, Moderate-to- treatment following is a cost-effective treatment with Etanercept societal 10 years QALY 2008; Swedish intermittent, and Adalimumab severe psoriasis interruption; treatment compared with for psoriasis perspective kronor/€ [57] 40 mg every other week hospitalisation cost; Adalimumab and non- cost of absenteeism, systemic standard of care and cost of unemployment, due to psoriasis Drug costs; costs of Economic evaluation of initial outpatient and Lloyd et al., Etanercept 50 mg biw, 12 weeks Etanercept in the UK National Moderate-to- follow-up outpatient Etanercept 50 mg biw is 2009; UK; 2006; Etanercept 25 mg biw, and no (extended to 10 QALY management of chronic Health Service severe psoriasis visits; costs of adverse cost effective in the UK £ [59] systemic therapy years) plaque psoriasis events, and inpatient days Calcipotriene + Corticosteroid Addition of the 308-nm (betamethasone), ICI Drug costs; costs of excimer laser to the (triamcinolone acetonide), office visits during Marchetti et al., Cost-effectiveness of Mild-to- rotational mix of Excimer Laser, UVB, PUVA Payer’s Treatment-free day, treatment and during 2005; USA; second line therapies of moderate 1 year treatments is expected to (UVA + methoxsalen capsules), perspective and remission day remission; costs of 2003; USD [30] psoriasis psoriasis add incremental clinical Anthralin + Corticosteroid management of adverse benefit for patients without (clobetasol), Tazarotene + events incremental cost for payers Corticosteroid (clobetasol) Acitretin, Alefacept, Cost-effectiveness and cost Pearce et al., Cyclosporine, Efalizumab, Physician office visits; Methotrexate appears to be of treatment failures Not Moderate-to- 2006; USA; Etanercept, Infliximab, 12 weeks PASI-75 nursing visits; and the most cost-effective associated with systemic mentioned severe psoriasis 2003; USD [36] Methotrexate, NB-UVB, and laboratory costs agent psoriasis therapies PUVA Direct (round-trip flight, roundtrip transfer to the hotel, hotel accommodation, Combined score from medical treatment, and Cost-effectiveness of Dead- (1) percentage of Shani et al., solarium fee), and Sea Climatotherapy versus Topical, phototherapy and Not Severe patients cleared, (2) Climatotherapy leads as 1999; Israel; 4-24 weeks indirect costs (loss of other modalities of systemic mentioned psoriasis length of treatment, the most cost-effective NR; USD [70] productivity during the treatment for psoriasis (3) mean remission, 4-week treatment plus and (4) annual cost 2-day flights, and treatment of possible side-effects, and the remission time) Sizto et al., Economic evaluation of Methotrexate, Cyclosporine, UK National Moderate-to- Not clear; Drug and associated Methotrexate and QALY 2009; UK; systemic therapies for Supportive care, Etanercept 25 Health Service severe psoriasis annualized monitoring and cyclosporine are cost

Author(s), Sources of Study Year; Economic question(s) Patient Cost components Comparators Perspectives Time horizon effectiveness Final decisions Location; Cost addressed characteristics considered evidence Year; Currency 2005/2006; moderate to severe psoriasis mg intermittently, Etanercept 50 perspective estimates administration costs; effective but require £ [62] mg intermittently, Efalizumab, presented cost of hospitalizations; monitoring for toxicities. Adalimumab, Etanercept, cost of productivity Of the biologics, Infliximab considered in Adalimumab was the most sensitivity analysis; cost effective justified for not using cost of adverse effect Acitretin, Alefacept, Adalimumab, Cyclosporine (5 Costs of medication; Staidle et al., Phototherapies and Cost-effectiveness of severe mg/kg/day), Etanercept Third-party Moderate-to- PASI-75 and DLQI office visits; laboratory 2011; USA; 1 year methotrexate are the most psoriasis therapy Infliximab, PUVA, payers severe psoriasis MID tests and monitoring 2010; USD [38] cost-effective options Methotrexate, NB-UVB, Home procedures NB-UVB, Ustekinumab Direct cost; Stern, 1988; Cost-effectiveness of Tar is NOT a cost- Not Cost to clearing (no transportation, parking, USA; NR; USD topical tar therapy in Topical tar vs. UVB Not mentioned Not mentioned effective option compared mentioned PASI or DLQI ) days of work lost, [39] psoriasis to UVB leisure days lost Drug cost; consultation Evaluate efficiency of fees; screening tests; Home-based phototherapy home-based phototherapy Vano-Galvan et Home phototherapy vs. Moderate-to- costs of phototherapy with narrow-band UVB with narrow-band UVB Payer's al., 2012; Spain; biologics (Etanercept, severe 16 weeks PASI-75 (costs of unit, delivery radiation was cost- radiation compared with perspective NR; € [47] Adalimumab, Infliximab) psoriasis; n=12 and collection of the effective compared with biologic drugs for the unit, consultations, biologic drugs treatment of psoriasis tests) as applicable * (1) TCF (Two Compound Formulation), first-choice, once daily for 4 weeks; followed by TCF, second-choice, once daily for 4 weeks; (2) Calcipotriol, first choice, once daily for 4 weeks; followed by potent steroid (BDP), second-choice, daily for 4 weeks; (3) Calcipotriol, first-choice, twice daily for 4 weeks; followed by potent steroid (BDP), second-choice, daily for 4 weeks; (4) Potent steroid (BDP), first-choice, daily for 4 weeks; followed by calcipotriol, second-choice, once daily for 4 weeks; (5) Concurrent calcipotriol, first-choice, once daily (morning) and potent steroid (BDP) once daily (evening) for 4 weeks; followed by the same regimen, second-choice, for a further 4 weeks † (1) Satisfactory response: overall clinical response of "cleared" or "marked improvement" by both the investigator and the patient, and the cosmetic acceptability of treatment rated as "excellent" or "good" by the patient; (2) Very satisfactory response: criteria for a "satisfactory response" with addition that no lesional or perilesional irritation was experienced by the patient; (3) Satisfied patient: "satisfactory response" with exception that only the patient's overall clinical response was evaluated; (4) Very satisfied patient: "satisfied patient" but who also experienced no lesional or perilesional irritation whilst using treatment. ‡ Degree of psoriasis was estimated separately for head (h), body (b), upper limb (u) and lower limb (l) for scaling (S), thickness (T) and area (A) of skin involvement. Scaling and thickness were scored from 0 to 3 (0: no signs, 1: slight involvement, 2: moderate involvement, 3: severe involvement) and skin involvement area was scored as percentage of the body area in question. The PSI was calculated as: 0.1 ∗ {[0.1퐴ℎ(푆ℎ + 푇ℎ)] + [0.35퐴푏(푆푏 + 푇푏)] + 0.2퐴푢(푆푢 + 푇푢)] + [0.35퐴푙(푆푙 + 푇푙)]}. The constants in the brackets represent the share of head, body, upper limbs and lower limbs in the total body area; the PSI can range from 0 to 60. NR= Not Reported; USD= United States Dollar; CAD= Canadian Dollar; £= British Pound Sterling; €= Euro; FIM= Finnish Markka; CHF= Swiss Franc; NHS= National Health Service; QALY= Quality-Adjusted Life Years; PASI= Psoriasis Area Severity Index; HRQOL= Health-Related Quality of Life; DLQI= Dermatology Life Quality Index; MID= Minimally Important Difference; UVB= Ultraviolet-B; NB-UBV= Narrow-band UVB; PUVA= Psoralen + Ultraviolet A (UVA) Phototherapy; PUVAsol= Psoralen + natural sunlight; ICI= Intralesional Corticosteroid Injections; b.d.= Twice Daily; eow= Every Other Week; BIW= Twice weekly; IM= Intramuscular; SC= Subcutaneous; IV= Intravenous.

Table 3: Summary of Sensitivity Analyses and Cost Drivers

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

TOPICAL The success rate of calcipotriol was 0.784 (instead of 0.608 in the baseline case) and the success rate of Success rate, drug cost per patient and doses Ashcroft et al., dithranol was 0.542 (instead of 0.496 in the baseline One-way No Cost and efficacy affected the ICERs but not the cost-effectiveness 2000 [48] case). Different average drug costs per patient for conclusions. calcipotriol and dithranol and different dose for dithranol The patients’ compliance remained as observed within the studies, i.e. with transition probabilities as Compliance with non-fix No impact on final conclusions: The fix Augustin et al., taken from the trials (maximum compliance of non- One-way No combination and effectiveness of combination either dominated the non-fix 2007 [49] fix combination). The effectiveness of the fix the fix combination combination or was more cost effective treatment. combination of calcipotriol/betamethasone varied by ± 10%. Compliance did not change the superiority of compound product. A 10% reduction in effectiveness of compound product did not change It assumed a maximum compliance for all treatment its cost-effectiveness status. The robustness of the arms. The effectiveness of the compound Patient compliance, effectiveness of model presented here was also supported by calcipotriol/ betamethasone product varied by ± 10%. Augustin et al., the compound calcipotriol/ further sensitivity analyses. Treatment with the The costs of calcipotriol and tacalcitol were adjusted One-way Yes Cost of tacalcitol 2009 [50] betamethasone product, cost of compound product was more cost-effective than until the cost-effectiveness of morning/evening calcipotriol and tacalcitol tacalcitol therapy; only at € 0.23/g or less did administration of calcipotriol and betamethasone or tacalcitol monotherapy become equally or more tacalcitol therapy was equal to that of the compound cost effective. Compared with calcipotriol, the calcipotriol/ betamethasone product. compound product remained more cost effective even if calcipotriol was free. Costs of phototherapy, amount of the TCF of calcipotriol and betamethasone dipropionate used (cost), baseline utility, utility on waiting list, PASI- The cost of phototherapy, cost of TCF (assuming Cost of phototherapy, TCF at 75 per treatment option, utility gain maximum possible use of drug), baseline utility Bottomley et al., maximal possible dose, baseline magnitude with response and non- Alternative values or plausible limits or 95% One-way Yes and utility on the waiting list had the most notable 2007 [51] utility and utility on the waiting response, response to phototherapy, confidence interval values influence on the model results. The variations of list relapse rate of the comparators, other parameters did not alter the conclusions. duration of waiting list, topical prescribed while awaiting phototherapy, use of potent steroid other than the least costly and most commonly used namely betnovate Colombo et al., No impact on the conclusions: Gel is better than One-way No Cost of dovobet gel The cost of dovobet gel varied from €0.65 to €0.69. 2012 [44] ointment. Marchetti et al., Drug acquisition cost, medical care The rank order was stable: A large change would The analysis assessed the degree of change needed to One-way No 1998 [31] cost, efficacy, length of treatment be necessary for tazarotene 0.1% to lose its first indicate a change to the next rank-ordered therapy.

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

rank and 0.05% to lose its second rank; Unrealistic levels would be needed to improve calcipotriene's ranking. The analysis was sensitive to changes in the cost and amount of Calcipotriol: Calcipotriol which Values of parameters varied by confidence intervals was cost effective at the base case could become in the meta-analyses or a range derived from panel dominant over fluocinonide and betamethasone Cost and amount of calcipotriol Success rate, cost and amount of discussions. The mean initial usage of calcipotriol Oh et al., 1997 dipropionate. The analysis was sensitive to One-way Yes and utility ratings of the side calcipotriol used, utility ratings of 30.6 g per week was used rather than the panel's [22] changes in utility ratings of the side effects of effects of fluocinonide the side effects of fluocinonide baseline estimate of 45 g per week. The decrement in fluocinonide: Fluocinonide could become the health status associated with fluocinonide approached dominant strategy. Variations in the values of that of calcipotriol (i.e., 0.02). other parameters did not affect the study conclusions. Papp et al., 2012 Probabilistic The deterministic model was stable and not No Cost and efficacy Simulations for probabilistic sensitivity analysis [64] sensitivity analysis sensitive to the uncertainties in application. Apply the non-observed period the mean cost assessed on the same treatment group for patients who completed the study follow-up, adjusted to the Peeters et al., Cost and study follow-up One-way No No impact was found on the conclusions. time left to spend until day 56. Cost-effectiveness 2005 [65] completion was estimated only for patients who completed the study follow-up, defined as patients for whom medical resource data were captured for 8 weeks. For Trunk/limbs psoriasis only, base-case conclusions within the framework of the restricted-comparator scenario were insensitive to changes in efficacy, adherence, cost and time horizon; To be a cost-effective first-line treatment option, threshold analyses showed that the unit Scenario analyses and a series of one-way sensitivity Clinical effectiveness, reduced One-way and cost of TCF product would need to drop by 55– analyses were performed to assess how changes in Sawyer et al., adherence, early vs. late response, probabilistic No 70% given perfect adherence and by 20–50% one or more parameters might change the 2013 [60] utilities, topical use, acquisition sensitivity analysis given reduced adherence. For scalp psoriasis, the conclusions of the analysis. Simulations for cost, referral, time horizon conclusions of were insensitive to changes in key probabilistic sensitivity analysis parameters of efficacy, adherence, cost and time horizon; A threshold analysis showed that only given a 60–70% discount would TCF product represent a more cost effective first-line treatment than potent corticosteroids alone. PHOTOTHERAPY Identical results using the SF-6D were found. Mean costs for home phototherapy became lower QALYs were calculated from SF-6D utilities instead than the costs for outpatient phototherapy. of EQ-5D utilities. The costs of absence from paid One-way, two-way, Utility values, including costs of Koek et al., 2010 Combined with the gain in QALYs for home work and unpaid work were calculated. The and probabilistic No work absenteeism, both treatment [58] phototherapy, the alternative calculation of costs treatment costs for both interventions were calculated sensitivity analysis costs using invoice prices would produce dominated strategy. Overall, the using invoice prices. Bootstrap resampling for conclusions did not change, i.e., the home therapy probabilistic sensitivity analysis was more cost effective.

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

SYSTEMICS TRADITIONAL SYSTEMICS Potential The relative effectiveness of cyclosporine to produce Relative efficacy of Relative effectiveness of The relative efficacy of cyclosporine and impact a year of clear skin varied over a range from Ellis et al., 2002 cyclosporine and methotrexate methotrexate and cyclosporine at methotrexate (i.e., the clearing rate) and the cost One-way depending on approximately 1 time to 20 times as effective as [27] (clearing rate) and cost of producing improvement and of cyclosporine were the only two variables that the threshold methotrexate. Costs varied over a wide range for a cyclosporine clearance, cost, rates of side effects had a significant impact on the results. chosen number of inputs. Cost varied because tapered discontinuation was less expensive than abrupt discontinuation and efficacy Hakkaart-van Multi-way and Clinical practice, healthcare system, No indication was found that the conclusions varied because tapered discontinuation produced Roijen et al., probabilistic No unit costs, efficacy, climate differ substantially between countries. more systemic therapy-free days on average than 2001 [68] sensitivity analysis abrupt discontinuation. Bootstrap resampling for probabilistic sensitivity analysis BIOLOGICS In a one-way analysis, the cost-effectiveness was sensitive to variations in average wholesale price Vary the efficacies of the therapies as measured by and the percentage of patients achieving PASI-75 DLQI and PASI-75 improvement and the total cost of or DLQI at a level of ± 5 % and 10%. Variation of treatment by ± 5% and ± 10% while holding all other Ahn et al., 2013 One-way and two-way Yes DLQI and PASI-75 Cost and efficacy the average wholesale price resulted in a values at their ‘best estimate’ or baseline value. [23] respective ± 5 % change in the cost-effectiveness Extreme case scenarios, both best and worst case: the ratio in all cases, and therefore, did not affect lowest cost and highest clinical efficacy vs. the which agents were most cost effective. The cost- highest cost and lowest efficacy effectiveness varied by efficacy. Rate of hospitalizations and assumptions regarding weight and waste of excess drug were found to be influential parameters: Changing the percentages of patients who were assumed to be hospitalized was influential to the ICERs; Changing the average length of stay from 17 to 19.5 days did not noticeably affect the results; Lowering the average patient weight from The study adjusted the length of stay to a greater that reported in the clinical trials of adalimumab estimate (19.5 days) and explored the scenarios (i.e. 93 kg) to that reported in trials of infliximab Frequency of hospitalization for Frequency and length of where 0% and 100% of non-responders were One-way and (90 kg) did not affect the most cost-effective Anis et al., 2011 non-responders, weight and hospitalization for non-responders, hospitalized. The assumptions on different weight probabilistic Yes sequence of treatments. Only when the average [24] waste of excess drug, average weight and waste of excess drug, (93kg to 90kg), excess wastage vs. no excess wastage sensitivity analysis weight was assumed to have been 90 kg and when wholesale price and dosage utility weight, drug cost, dosages assumptions, UK weights for utility, average excess drug was assumed to have been shared wholesale price and including dosages other than among patients did infliximab advance to the recommended second treatment in the sequence. However, the rest of the sequence remained similar to the base case. Using average wholesale price instead of weighted average costs increased all costs and, in turn, increased all ICERs. Including dosages other than those recommended in product labels changed the sequence of treatments due to the differences in the drug costs.

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

Baseline scenario, best-case scenario, worse-case The ratio varied by both efficacy and cost but the Blasco et al., scenario: cost for weight-dependent dosing and Two-way No Cost and efficacy of biologic agents sensitivity analysis confirmed the robustness of 2009 [41] efficacy 95% confidence interval upper and lower these figures and the conclusions did not change. limits Potential The wholesale price of infliximab varied by ± 10%. impact Price of infliximab, cost and de Portu et al., Price of infliximab, cost and ICERs were quite sensitive to price variations of The best and the worst possible scenarios: the lowest One-way and two-way depending on efficacy (PASI-75) of all 2010 [46] efficacy infliximab and best case and worse case scenarios. cost and highest efficacy vs. the highest cost and the threshold treatments lowest efficacy chosen Base case and the most and least favourable scenarios for each treatment: For infliximab, the most favourable and least favourable estimates were calculated using the 95% confidence intervals for the The advantage of adalimumab was difference in weight between subjects with and Ferrandiz et al., Two-way No Cost and efficacy methodologically robust because it was the most without psoriasis. When drug was given in a single 2012 [52] cost efficient in the three scenarios. dose (e.g. adalimumab, etanercept, ustekinumab), the costs for the three scenarios were the same. The most and least favourable estimates of incremental efficacy were calculated using the upper and lower limits of the 95% confidence intervals, respectively. The variation of total treatment costs had a strong impact, changing the response rates of the initial treatment had a moderate impact, while varying the response rates of the treatment change had a Sensitivity analyses comprised a 25% and a 50% Greiner & relatively small impact on the ICERs at 36 weeks. reduction of PASI-75 response for the treatment Braathen, 2009 One-way No Cost and efficacy However, these analyses did not substantially change of non-responders and the variation of total [54] affect the outcomes of the cost-effectiveness treatment costs as well as of the response rates for the model and confirmed the robustness of the model initial therapies by ± 25%. in all cases, with lowest ICERs for infliximab and adalimumab followed by higher ratios in the ranking efalizumab, etanercept, and alefacept. The sensitivity analysis with non-weighted ustekinumab 45 mg was comparable with the base Efficacy for ustekinumab 45 mg was changed to non- Non-weight-based efficacy for case analysis, which confirmed that ustekinumab Drug costs estimated by the weight-based PASI-75 response rate. Drug cost for Igarashi et al., ustekinumab, drug costs estimated 45 mg remained the lowest cost per responder. One-way Yes number of gender-specific vials infliximab was changed based on assumptions that 2013 [71] by the number of gender-specific The cost per responder for infliximab decreased necessary for infliximab male patients used four vials and female patients used vials necessary for infliximab due to the reduction of the annual drug cost and three vials. showed better cost efficacy than adalimumab for Year 2. For the week-52 analysis, PASI-75 response rates were carried forward from week 12 to week 52 for All sensitivity analyses had no impact on Time horizons, PASI-90 response both ustekinumab and etanercept. The cost per Martin et al., conclusions: The cost per responder for One-way No rates, weights impact on responder at week 16 was evaluated using PASI-90 2011 [32] ustekinumab remained lower than that for ustekinumab cost response rates. Alternate weight distributions etanercept. (proportion of patients > 100 kg) were analyzed for ustekinumab. Menter and One-way Potential Median treatment-free response Median treatment-free response A shorter period adversely affected cost-efficacy, > 7 months treatment-free response period for

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

Baker, 2005 [21] impact (the period for alefacept and period for alefacept, etanercept whereas longer off-treatment response periods alefacept, >12 weeks of the higher doses of impact etanercept high-dose period high-dose period, duration of enhanced the cost-efficacy of alefacept. If the etanercept therapy, longer duration of efalizumab results not efalizumab therapy etanercept high-dose (50 mg) period was extended therapy shown) beyond 13 weeks, the cost-efficacy of etanercept decreased. Cost efficacy decreased with a longer duration of efalizumab therapy. The cost-effectiveness ratios were sensitive to variations in average wholesale price and DLQI and PASI-75 efficacies at a level of ± 5% in the Efficacies of the therapies: both The variables were varied by factors of ± 5% and ± Average wholesale price, costs one-way sensitivity analysis. In analysis of Nelson et al., DLQI and PASI-75, average 10%. Best-case and worst-case scenarios: the lowest One-way and two-way Yes and efficacies (DLQI and PASI- extreme scenarios, significant overlap in multiple 2008 [34] wholesale price of medications, cost and highest efficacy vs. the highest cost and 75) agents was observed at the 5% variance level. total physician costs, total lab costs lowest efficacy There was significant overlap in the cost- effectiveness ratios of all of the agents (DLQI) or multiple agents (PAIS-75) at 10% variance level. Sensitivity analyses did not affect the conclusions: Discontinuation rate had no effect on the model Discontinuation rate varied between 0 and 90%. outcomes. Reducing the model’s time horizon to 2 Time horizons were reduced to 2 or 5 years. Discount One-way and Discount rate, time horizon, or 5 years did not change the ultimate outcome of Pan et al., 2011 rates were 0% and 3%. The utility values derived probabilistic No discontinuation rate, utility values ustekinumab dominating etanercept. Similarly, [35] directly from EQ-5D scores from a different source sensitivity analysis for PASI scores changing the discount rates did not change the were used. Simulations for probabilistic sensitivity outcome. The utility gain with ustekinumab analysis versus etanercept was greater with this parameter change and thus ustekinumab remained dominant. Cost-effectiveness ratios were sensitive to variation of both, the PASI-75 efficacy parameter and the cost in the one-way sensitivity analysis at One-way sensitivity analysis and extreme scenario a level of 5%. Especially varying PASI-75 analysis: here all variables were varied efficacy led to a marked effect. This effect was simultaneously to determine a ‘best-case scenario’ most distinct in the agents showing ‘low’ efficacy. Schmitt-Rau et and a ‘worst-case scenario’. PASI-75 improvement One-way and two-way Yes Cost and efficacy (PASI-75) Cost and efficacy Cost-effectiveness ratios of the ‘high-efficacy’ al., 2010 [61] and pharmacy retail price were varied by a factor of ± drugs such as infliximab 5 mg/kg were barely 5%. This factor was chosen on the basis of the 95% influenced. Marked overlap of cost-effectiveness confidence intervals of the pooled efficacy data from ratios was observed when cost and efficacy were the randomized controlled trials. varied at a 5% level in the extreme-scenario analysis, indicating differences observed may not be true differences. Utility values, self-administration of etanercept 50 mg, These results were robust to sensitivity analyses discontinuation rate, drug price, with few exceptions: only when 100% patients time horizon, mortality, PASI-50 administer etanercept 50 mg, ustekinumab 45 mg One-way and Villacorta et al., response rate for ustekinumab 90 did not dominate etanercept 50 mg but was cost- A probabilistic sensitivity analysis explored the base- probabilistic Yes Time horizon 2013 [40] mg, healthcare costs for patients in effective; ustekinumab 45 and 90 mg were not case model uncertainty with simulations. sensitivity analysis the PASI<50 health state, lost time cost effective for up to a 12-week time horizon for ustekinumab and etanercept compared with etanercept 25mg weekly and 50mg treatments, hourly compensation, weekly. discount rate, starting doses of

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

etanercept Scenario I: fourth quartile DLQI at baseline and no The ICERs of biologicals decreased when hospitalization for non-responders; Scenario II: any assuming the fourth quartile DLQI at baseline or DLQI at baseline, 21 days inpatient hospitalization Probabilistic Baseline utility value assessed Baseline utility value assessed by 21-day hospitalization for non-responders, when not responding to therapy; Scenario III, fourth Woolacott et al., sensitivity analysis for by baseline DLQI and Yes baseline DLQI, hospitalization for separately or altogether. When combining the two quartile DLQI at baseline and 21 inpatient days; 2006 [67] four different hospitalization for non- non-responders assumptions, the ICERs of biologicals were the Scenario IV: comparison of biologicals with other scenarios responders lowest indicating they enter a cost-effective systemic therapies (any DLQI at baseline and 21 sequence at the lowest threshold. inpatient days); Simulations for probabilistic sensitivity analysis MIXED Goeckerman efficacy did not alter its ICER to no therapy but slightly changed the ICER to Methotrexate efficacy (clearing Efficacy of both Goeckerman and methotrexate. Methotrexate efficacy did not alter Base case, best case and worst case scenarios: the Chen et al., 1998 One-way Yes rate) and utilities at different methotrexate, their costs, and its ICER to no therapy but ICER of Goeckerman best plausible assumption vs. the worst plausible [26] severity levels utilities to methotrexate was sensitive to methotrexate assumption efficacy. ICERs were sensitive to variations in utilities at different severity levels. Cost of etanercept had more impacts on ICERs. Potential ICERs did not differ significantly from the values impact Cost of etanercept, cost of basal Colombo et al., Cost of etanercept and efficacy obtained in the base case, in both groups of One-way depending on treatment, cost of hospitalization, All estimates varied by ± 20%. 2009 [43] (PASI) of etanercept disease severity when cost of basal treatment and the threshold efficacy and discounting rate hospitalization cost changed. Changes in efficacy chosen moved the ICER by 3% to 12%. Methotrexate remained the least costly/treatment success when a range of efficacies was Costs varied by a range of doses (etanercept at 25 or considered. UVB and PUVA had similar costs; 50 mg twice-weekly, infliximab at 5 and 10 mg/kg) Cyclosporine was more than phototherapy but less Feldman et al., and modalities (alefacept intramuscularly or One-way No Doses, modalities, efficacy than biologicals; Acitretin and infliximab at 2003 [28] intravenously). High and low efficacies were plugged 5mg/kg dose had similar costs when sensitivities into the model and were reported as negative and were considered. All of other biologicals had positive error bars respectively for each treatment. comparable costs, remaining much more costly than any of the other treatments. Choice of topical treatments in the reference pathway and all four For example, it switched the reference pathway (now treatment pathways, percentage of TCF first- and second-line) and treatment pathway 1 Changing the reference pathway to potent steroid Freeman et al., patients progressing through each (now calcipotriol twice daily (first-line) and potent Multi-way No (first-line) and calcipotriol twice daily (second- 2011 [53] treatment pathway, the magnitude steroid (second-line) and assumed that treatment line) had negligible impact on the results. of efficacy or costs and the option pathways 1, 2, 3 and 4 typically had 45%, 25%, 25% to include a new (or not commonly and 5% patients, respectively. used) topical Price of the hourly wage for a The price of the hourly wage for a dermatologist, the One-way and dermatologist, cost price per visit to cost price per visit to the UVB, the cost price for Hartman et al., The impact of changes in cost prices on the mean probabilistic No the UVB, the cost price for short short contact treatment, and the price for stay in 2002 [55] total cost per patient was limited. sensitivity analysis contact treatment, price for stay in hospital were varied; Charges, calculated costs in a hospital, number of clearance days teaching hospital, and the calculated costs in a

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

general hospital were used; The lowest and the highest values of these three prices were included in the sensitivity analysis. Bootstrap resampling for probabilistic sensitivity analysis It was shown that with increasing costs of basal treatment, the incremental costs of etanercept treatment declined. Altogether, the model proved Heinen- Cost of basal treatment, cost of to be relatively stable. The assumed hospital costs Treatment costs for basal treatment varied between Kammerer et al., One-way No etanercept, probability of a hospital had the greatest influence on results. By reducing €20 and €120 per cycle. 2007 [56] stay the probability of a hospital stay by 30 %, the costs per QALY rose only moderately by 24,036 €/QALY. Overall, the conclusions remained. Altering the discounting rates affected the ICERs Alter discounting rates between 0% and 5%. All but not the conclusions. One-way sensitivity resource costs, hospitalization costs and cost due to Discount rates for cost and health analysis of the resource and indirect costs had loss of working hours (indirect costs) varied within a One-way and Utilities for the PASI-75 health Knight et al., effects, resource costs, little effect on the ICERs. This was also true of range of ± 25% from the base-case value. Alter the probabilistic Yes state and the PASI-50–74 health 2012 [57] hospitalization costs, indirect cost, the efficacy rates. Altering the utility values for utility values for the PASI-75 health state to its 95% sensitivity analysis state efficacy rates, utility values the PASI-75 health state and the PASI-50–74 credible interval of 0.986 and reduce the utility for health state had the largest effect on the the PASI-50–74 health state to its 5% credible deterministic results. interval value of 0.751. Baseline PASI and DLQI scores, the treatment-free period between The cost-effectiveness of both etanercept cycles of therapy, the percentage of regimens was sensitive to the assumed the response achieved on re- requirement for hospitalization in untreated One-way and treatment, the number of hospital Each variable varied within a range of values. Lloyd et al., 2009 Response rate achieved in re- individuals and to the response rate achieved in probabilistic Yes days for patients who did not Bootstrap resampling for probabilistic sensitivity [59] treatment re-treatment. The analysis was found to be less sensitivity analysis respond to treatment, the number of analysis sensitive to the frequency of clinic visits required outpatient visits required per cycle for monitoring or to the method of discounting of treatment, the duration of applied to future costs and benefits. therapy before treatment interruption, discounting rates Pearce et al., The efficacies of the agents varied by a factor of ± One-way No Efficacy No impact was found on the conclusions. 2006 [36] 5%. Changing the number of days hospitalized owing to nonresponse to treatment had an important effect: It increased the cost-effectiveness of all Hospitalization days for non- treatments when the number of days was responders, disutility on increased from 21 (base case) to 39 and all intermittent therapy, dosage of Uncertainty was assessed using both a one-way treatments were considered cost effective at £30 One-way and Hospitalization days for non- cyclosporine, continuous sensitivity analysis where key parameters are Sizto et al., 2009 000 threshold; When it was assumed that non- probabilistic Yes responders and lost productivity cyclosporine use, dosage of sequentially changed to alternative plausible values [62] responders did not require hospitalizations, the sensitivity analysis while hospitalized etanercept, utility values, PASI-50 and a probabilistic sensitivity analysis. Simulations ICERs of methotrexate and cyclosporine became response, weight, lost productivity for probabilistic sensitivity analysis positive. Including the costs of lost productivity while hospitalized, proportion of during hospitalization increased the cost- non-responders hospitalized effectiveness of all treatments because of the avoidance of this added cost to society and all treatments except infliximab were considered cost

Impact on Author(s), year Sensitivity analysis Key drivers Uncertain parameter estimates Detailed impact of uncertainty Details of sensitivity analysis conclusions

effective at the threshold. The variance of other variables did not change the conclusions. The robustness of our results was demonstrated by Base case, best case and worst case scenarios: a 15% Vano-Galvan et the sensitivity analysis, which showed home Two-way No Cost and efficacy variation in the critical variables of effectiveness and al., 2012 [47] phototherapy to be more cost-effective than cost biologic therapy in all 3 scenarios. Note: Blank cells indicate ‘not applicable’. ICER: Incremental Cost-Effectiveness Ratio; TCF: Two Compound Formulation; SF-6D: Short Form-6D; EQ-5D: EuroQol-5D; QALYs: Quality-Adjusted Life Years; PASI: Psoriasis Area Severity Index; DLQI: Dermatology Life Quality Index; UVB: Ultraviolet-B; PUVA: Psoralen + Ultraviolet A (UVA) Phototherapy